Multi-physics optimization of a smart actuator for an automotive application

Florent Robert (1, 2, 3) , Filipe Vinci dos Santos (3) , Christophe Gutfrind (2) , Laurent Dufour (2) , Philippe Dessante (1)

1 GeePs - Laboratoire Génie électrique et électronique de Paris
2 EFi Automotive - Service Recherche Avancée
3 Chaire Conception Analogique Avancée Thales/Supélec

This paper describes a design framework for optimization of automotive mechatronic systems that uses analog simulations and evolutionary algorithms. We demonstrate the effectiveness of our methodology in sizing a complex actuator under multi-physic constraints. Our challenge was to design a compact exhaust gas recirculation (EGR) valve meeting the severe electrical, thermal, mechanical and magnetic requirements of the system.

Mots clés: «Design»; «Mechatronics»; «Automotive application»; «Thermal design»; «System engineering»

Domaines: Energie électrique; Electronique; Sciences de l'ingénieur [physics]

Florent Robert, Filipe Vinci dos Santos, Christophe Gutfrind, Laurent Dufour, Philippe Dessante. Multi-physics optimization of a smart actuator for an automotive application. 2015 17th European Conference on Power Electronics and Applications (EPE'15 ECCE-Europe) , Sep 2015, Genève, Switzerland. ⟨10.1109/EPE.2015.7309129⟩. ⟨hal-01254975⟩